Vuk Lesi

Important changes to forums and questions

All forums and questions are now archived. To start a new conversation or read the latest updates go to forums.mbed.com.

7 years, 8 months ago.

LPC1768 revisions

We have an experimental setup with custom made PCBs that have headers for mbed LPC1768 boards. When we ordered more mbeds obviously a newer series arrived with slightly different components (capacitors are different color, the reset pushbutton is different, MCU date and batch codes are different, of course, and so on...). Now the problem is that the new boards do not work right with the same code in our setup with Microchip MRF24J40MA radios. Reception is very bad and is effected by hovering your hand over the PCBs. The attached photo shows both boards (old on the left and new on the right):

/media/uploads/vuklesi/imag01574.jpg

Has anyone else encountered any similar problems or does anyone have any suggestions?

Question relating to:

Last updated: 30 Jun 2020
Rapid Prototyping for general microcontroller applications, Ethernet, USB and 32-bit ARM® Cortex™-M3 based designs

1 Answer

Wim Huiskamp
7 years, 8 months ago.

This is probably EM interference. The LPC1768 runs at about 100 MHz and has just a single GND pin. There were reports of LPC1768 modules interfering with GPS receivers as well. You might try shielding the mbed using some metal cage or foils. Also look at the powersupply. You might need some LC filtering on the VCC and GND lines that are shared between the radio and the mbed.

Accepted Answer

Thank you for your answer. I haven't been very successful with shielding the mbed. I tried putting the radio on long wires to move it away from the MCU and also experimented with separate power supplies, but nothing really worked. What bothers me most is that the old mbeds work fine, and newer ones don't.

Thanks

posted by Vuk Lesi 11 Aug 2016

Since I cannot formally answer my own question, I'll post my findings as a comment. I just found a solution for this problem. The reason why transmissions were failing was Clear Channel Assessment. For some reason the new mbed effects the Received Signal Strength Indicator read by the radio. The threshold was set to minimum thus the channel was considered busy until you start touching the board when some type of noise gets settled. I changed the radio configuration so that a more noisy channel is acceptable. This solved the problem. I don't understand why would the mbed board effect the signal strength reading on 2.4GHz, but it is probable that something else on the radio module is actually effected which manifests on the RSSI reading. I will once again emphasize that this issue is not present under the same old configuration with an older batch of LPC1768 mbeds that we have.

Thanks

posted by Vuk Lesi 12 Aug 2016